Almost two decades ago I saw in the arcades the futuristic fighting game “Rise of the Robots”. As a youngster I was imagining what the future of computing and robotics could be. The game ended up not being that great, regardless of that it wasn’t visually very realistic, but instead, it relied on the gameplay and partly on the player’s imagination for the immersion. Hence, around this time, I was dreaming of tablets (from Star Trek) and completely autonomous robots that would help us with everyday tasks (like terminators, without the killing part of course and maybe the Jetson’s robots?).
Yet the future was not exactly what I was expecting. After all this experimentation and technological progress, it seems that people hyped with Chatbots (or chatterbots) instead!
What exactly is a Chatbot?
A chatbot is a computer program that can provide intelligent conversation. Currently chatbots focus on providing several automations and accepts input via a chat interface. Facebook recently, in the f8 developer conference (https://www.fbf8.com/) introduced such functionality in the app “facebook messenger”. In addition, Slack already permits its users and potential developers to create such bots for quite some time now, by just using its APIs.
Actually this whole chatbots theme, which are heavily backed up by AI (artificial intelligence), reminds me the Turing test (https://en.wikipedia.org/wiki/Turing_test), which developed back in the 1950s (Turing, Alan (October 1950), “Computing Machinery and Intelligence”, Mind LIX (236): 433–460, doi:10.1093/mind/LIX.236.433). Exactly for this reason, all this chatbot hype felt to me a little backwards to me, at least in its current state.
One Step Forward and Two Steps Back
Many people around me are very excited, and lot of them implemented chatbots and claim that they are “intelligent”, because they are backed up by some sort of an AI. But as far as I have seen, they are not. Or at least, the implementations that are public domain right now are just scratching the surface of what a chatbot could be.
In addition, Marketing guys are talking about conversational e-commerce, which will let you perform shopping through an textual interface. I understand that if it is implemented right, it has some aspects of the problems solved, e.g. user identity, payment methods, etc. The truth is: most of these systems right now are more prototypes, than usable programs. It seems to me, that by making people type specific commands that they look like natural language, but they aren’t, developers try to bypass (for now) the voice recognition problem, which prohibits them to focus on the AI and the functionality, since they have to make it work first. Or as I stated in my previous article “The Revenge of Command-line”, is the sweet revenge of Unix lovers.
So We Wait, The Future is not Here Yet
Or I am not sure that all this will end up fulfilling our hopes and dreams, which are mostly in the direction of us communicating better with computers, automating our every day life. When I think of a killer program, it would be something like Apple’s Siri or Microsoft’s Cortana, or Amazon Echo that I could instruct something like “fetch me my usual coffee at home” and it should automatically notify the local cafeteria that I am a frequent customer that I want my cappuccino, right now. In addition, it should take care of the bill for me without asking me for permission. Something like the Amazon’s dash button, but for everything.
So all we have to do is wait. We have a long (and exciting) road in front of us. I am not sure where it will actually lead, but I think the next paradigm shift is slowly forming itself. The previous one was the absolute domination of the mobile phones. Companies tried to repeat it with the tablets, but it seems they have failed. So, the stake now is on this, Chatbots and how AI can simplify our day-to-day interaction with computers. But the AI is not there yet. Yes we can make autonomous cars and actually solve problem that focus on a specific problem domain, but we are not even close to general-purpose implementations.
Standing on the Shoulders of Giants
Things are getting more and more mature to support that shift. Scaling is a problem that is more or less taken care of, it is transformed into money, you just have to pay your cloud provider. A second wave of services are begin developed, with APIs that support computer vision, prediction and big data analytics, which are enablers for the next-level application to be developed. Which are going to be? A Siri that actually can understand my “Hellenic-English” accent? Maybe. When it does, I know that I will drink my cappuccino with less hassle at last.